[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Questions about -st- commands

From   "Miguel Nathan Foguel" <>
To   <>
Subject   st: Questions about -st- commands
Date   Fri, 20 Jul 2007 15:21:53 -0300

Dear all,
I want to study the difference in time to promotion of men and women. I
have measured promotion using the occupation of individuals at the
establishment they work, where promotion means being in a managerial
occupational category. I have a sample in which workers are followed
since they are hired by the establishment. One aspect of this sample is
that many individuals are already hired as managers, a fact that implies
that their failure time is zero. (Importantly, there is a significant
difference between men and women for this group.)When I try to -stset-
the data, those "zero-duration" observations are not set at all (their
_d, _t and _t0 are all set to missing). Does anyone know if there is a
way to -stset- these data preserving the zero failure time? If so, would
-streg- allow NON-proportional hazard estimation without taking logs of
(zero) durations? Thanks a lot in advance. 

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index